The present invention relates generally to networks, more particularly, to dynamically assigned Internet Protocol (IP) address networks, and even more particularly to the use of user-based policies in networks.
In a network, a policy-based management system maintains policies or rules that govern the use of or access to a network service. As used herein, a policy is a single rule which defines conditions that when evaluated true trigger actions to allow or deny the service. A number of policies can be combined together to form a policy group. However, a recent evolution in terminology of the art (not universally accepted and not followed herein) uses the term xe2x80x9cpolicyxe2x80x9d itself to mean the combination of more than one rule, and the term xe2x80x9crulexe2x80x9d to mean a single rule.
Previous methods for implementing policies in such systems have relied upon having fixed network addresses. Modern networks, however, more and more depend upon dynamic assignment of addresses for items attached to the network. In computing environments where network addresses are dynamically assigned to computers as they connect into the network, a user""s workstation or laptop computer no longer maintains a static network address, and often it does not maintain a hostname that is recognized by the computing environment. This is especially true when dialing into a corporation""s network using remote access mechanisms.
Previous solutions have also depended upon assigning policy implicitly based upon characteristics of a device or logical entity which is configured separately from the policy management tools. Such techniques lack flexibility in assignment of policy and lack centralized distribution to the network services being managed. In addition, previous proposed solutions do not resolve conflict between different functions on a manageable entity between policies with different action or condition types applied with a single rule. In fact, to date organizations that define standards for implementing policy have only loosely defined methods for associating policy with a managed entity.
Thus, there is a need for grouping policy related processes and resources, referred to herein as targets.
As networks have become more and more complicated, so has the management of those networks. The present patent document discloses novel methods and means for using rules that control interactions of entities in electronic systems, such as networks. A collection of such rules are referred to herein as policies. A network comprises processes and resources that provide services to other processes and resources which, in turn, are also connected to the network. In representative embodiments, the present document discloses techniques for grouping policy related processes and resources, referred to herein as targets.
As indicated, electronic systems, such as networks, that comprise resources or processes can control the interactions of such items by means of rules or policies. These items could be for example processes, functions, abstract objects, or physical electronic devices such as computers, printers, etc. Thus, policy refers to the description of a behavior or action that is desired for the item to which the policy applies. In network systems, policies are typically associated with items that affect the flow of data on that network. In order to affect that network traffic flow, policies are directed toward or targeted at managed or controlled entities. An example of a policy could be xe2x80x9cassign priority 5 to traffic from the user whose name is user_onexe2x80x9d.
As referred to herein, a target is a process or resource that is being managed using a policy or policies. The managed item itself may be able to recognize and conform to the policy, or may be managed by a proxy which recognizes policy information and converts it to configuration information that the managed entity can recognize and conform to.
Modern network devices are typically managed as a unit, i.e., the various features of the device are all managed together. For example, a router has multiple interfaces, with each interface representing a connection to one or more networks. The router""s function is to route traffic between these networks. Further, each interface can have multiple capabilities, each of which can affect the traffic in different ways. These mechanisms can each be configured separately. But, in modern network devices all of these different aspects of a single device are typically managed together, usually presenting a difficult to understand interface to the administrator of the network. As a result, the management of even a single device can become a daunting task. In representative embodiments, the present patent document discloses techniques by which separate aspects of a given device can be managed individually by policies.
An advantage of the representative embodiments as described in the present patent document is that the grouping of targets allows the administrator to easily view and manage the entities, whether logical or physical, that are involved in the delivery of a service which could be for example a database, access to a system, or some other service, together rather than individually.
The policy creator benefits from a single, consistent resolution mechanism for the policy-managed environment. Developers of client programs are relieved of the burden of providing for the name resolution themselves, they rely on the server program to perform this service. Central mapping also ensures that consistent information is used throughout the managed environment. Policies can now work in a dynamic environment with automated updates of the changing information without further intervention by the administrator, and with minimal effort on the part of the policy enforcement implementor. The server program would interact with the user name to network address mapping program to determine when an address is assigned and then notify the Policy Enforcement clients, the client programs, that a change had occurred, and what the new mapping is.